Textbooks play a fundamental role in the learning process: to encourage the acquisition of new knowledge, they typically introduce the fundamental concepts before the advanced ones. Such an ordering of concepts is reflected in prerequisite relations, e. g. in dependency relations indicating which concepts a student must know before acquiring a new one. During the seminar the main challenges related to the recognition of prerequisite relationships starting from the content of didactic texts will be discussed. PREAP, a protocol for the construction of annotated resources enriched with prerequisite relationships, and a strategy for the automatic recognition of these relationships in didactic texts will be then introduced. The seminar will be concluded with an overview of the research issues still open and the illustration of PREAP potential in addressing such challenges.
Speaker(s): Chiara Alzetta
She has been a research fellow since 2021 within the ItaliaNLP Group of the Institute for Computational Linguistics “A. Zampolli” of the National Research Council, to which she has been associated since 2018. In 2021 she received a PhD in Digital Humanities at the University of Genoa with a thesis on manual annotation and automatic recognition of prerequisite relationships among didactic concepts. Previously she receivedd a Degree in Digital Humanities at the University of Pisa with a thesis aimed at the study of linguistic types and universals based on a linguistic-computational approach. Her research interests concern the definition of strategies for the personalization of didactic paths and the exploration of language resources annotated.